Application of the K-Nearest Neighbor Method to Classify Milk Quality

Abstract

Milk is a food ingredient that is often consumed because it has many good benefits for the body. Fresh milk production during 2022 will be 129.98 million liters or 4.33 million liters per company. Milk production fell 2.39% from 2021. Monitoring to test the quality of cow's milk is very necessary. So it can produce dairy products of good quality. This research aims to classify the quality of milk using data mining techniques. The data that will be used includes physical and chemical parameters of milk. The classification method that will be applied is method K-Nearest Neighbor so you get the pattern you want to use. The public data used was 1059 data. The K value used for this research is K=1. The results of this research show that the K-NN method succeeded in achieving an accuracy of 94%. Based on the analysis results, the KNN method with a value of K=1 is an effective method in classifying milk quality with a curation of 94%, indicating that the resulting model is able to differentiate between high and low quality milk with a relatively small error rate. It is hoped that this research will help farmers improve milk quality and provide valuable information for the dairy industry in adopting more effective quality control systems